Benature/omnivore
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Escape HTML entities in puppeteer-parse

Browse Source
This commit is contained in:
Hongbo Wu
2022-09-23 16:40:32 +08:00
parent aef83ee958
commit cb609d893e
4 changed files with 11 additions and 6 deletions
Download Patch File Download Diff File Expand all files Collapse all files

3
packages/content-fetch/package.json
View File

@ -10,7 +10,8 @@
"jsonwebtoken": "^8.5.1",
"linkedom": "^0.14.9",
"luxon": "^2.3.1",
"puppeteer-core": "^16.1.0"
"puppeteer-core": "^16.1.0",
"underscore": "^1.13.4"
},
"scripts": {
"start": "node app.js",

1
packages/puppeteer-parse/package.json
View File

@ -14,6 +14,7 @@
"linkedom": "^0.14.9",
"luxon": "^2.3.1",
"puppeteer-core": "^16.1.0",
"underscore": "^1.13.4",
"winston": "^3.3.3"
},
"devDependencies": {

5
packages/puppeteer-parse/twitter-handler.js
View File

@ -6,6 +6,7 @@
require('dotenv').config();
const axios = require('axios');
const { DateTime } = require('luxon');
const _ = require("underscore");
const TWITTER_BEARER_TOKEN = process.env.TWITTER_BEARER_TOKEN;
const TWITTER_URL_MATCH = /twitter\.com\/(?:#!\/)?(\w+)\/status(?:es)?\/(\d+)(?:\/.*)?/
@ -116,7 +117,7 @@ exports.twitterHandler = {
const tweetData = (await getTweetById(tweetId)).data;
const authorId = tweetData.data.author_id;
const author = tweetData.includes.users.filter(u => u.id = authorId)[0];
const title = titleForAuthor(author)
const title = _.escape(titleForAuthor(author))
const authorImage = author.profile_image_url.replace('_normal', '_400x400')
let text = tweetData.data.text;
@ -157,7 +158,7 @@ exports.twitterHandler = {
<meta property="og:image" content="${authorImage}" />
<meta property="og:image:secure_url" content="${authorImage}" />
<meta property="og:title" content="${title}" />
<meta property="og:description" content="${tweetData.data.text}" />
<meta property="og:description" content="${_.escape(tweetData.data.text)}" />
</head>
<body>
${front}

8
packages/puppeteer-parse/youtube-handler.js
View File

@ -5,6 +5,7 @@
/* eslint-disable @typescript-eslint/no-require-imports */
require('dotenv').config();
const axios = require('axios');
const _ = require("underscore");
const YOUTUBE_URL_MATCH =
/^((?:https?:)?\/\/)?((?:www|m)\.)?((?:youtube\.com|youtu.be))(\/(?:[\w-]+\?v=|embed\/|v\/)?)([\w-]+)(\S+)?$/
@ -36,11 +37,12 @@ exports.youtubeHandler = {
const oembedUrl = `https://www.youtube.com/oembed?format=json&url=` + encodeURIComponent(`https://www.youtube.com/watch?v=${videoId}`)
const oembed = (await axios.get(oembedUrl.toString())).data;
const title = oembed.title;
const title = _.escape(oembed.title);
const ratio = oembed.width / oembed.height;
const thumbnail = oembed.thumbnail_url;
const height = 350;
const width = height * ratio;
const authorName = _.escape(oembed.author_name);
const content = `
<html>
@ -49,12 +51,12 @@ exports.youtubeHandler = {
<meta property="og:image:secure_url" content="${thumbnail}" />
<meta property="og:title" content="${title}" />
<meta property="og:description" content="" />
<meta property="og:article:author" content="${oembed.author_name}" />
<meta property="og:article:author" content="${authorName}" />
</head>
<body>
<iframe width="${width}" height="${height}" src="https://www.youtube.com/embed/${videoId}" title="${title}" frameborder="0" allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture" allowfullscreen></iframe>
<p><a href="${url}" target="_blank">${title}</a></p>
<p itemscope="" itemprop="author" itemtype="http://schema.org/Person">By <a href="${oembed.author_url}" target="_blank">${oembed.author_name}</a></p>
<p itemscope="" itemprop="author" itemtype="http://schema.org/Person">By <a href="${oembed.author_url}" target="_blank">${authorName}</a></p>
</body>
</html>`